Research On Deployment Method Of Post-disaster UAV Base Station Based On Swarm Intelligence

After earthquakes,floods,tsunamis,hurricanes,fires and other natural disasters or terrorist attacks occur,ground communication facilities are in danger of being destroyed.With the rapid development of unmanned aerial vehicle(UAV)technology,UAV can be used as an air base station to support User Equipment(UE).UAV-base station(UAV-BS)has the advantages of fast moving speed,high flexibility and easy control.It can be used for rapid deployment of emergency communication networks.Therefore,the deployment of UAV-BS plays an important role in rescue operations and post-disaster reconstruction.However,the current post-disaster UAV base station deployment is still in its infancy.There are still many challenges in the research on optimization methods for post-disaster UAV base station deployment.First of all,how to maximize the network throughput by deploying UAV-BS reasonably while maintaining the quality of service(QoS)is a challenging problem.Secondly,how to maximize the network throughput and the number of UEs covered by UAV-BS in a Device-to-Device(D2D)post-disaster scenario by deploying UAV-BS reasonably is another challenging issue.The D2D-content delivery network(D2D-CDN)can forward the content received from the base station to other UEs that are not covered by the base station in a multi-hop manner.This network is widely used because it can greatly expand the coverage of base stations.However,infrastructure such as ground base stations are destroyed in disaster scenarios.Due to the lack of base stations as content sources,D2D-CDN cannot operate normally.UAV can be used as a temporary content source to flexibly deliver content to terrestrial UEs.However,in the post-disaster scenario for D2D-CDN,the deployment of UAV-BS needs to consider the transmission mode between D2D,such as single-hop transmission and multi-hop transmission.Finally,how to build a UAV-BS deployment optimization model suitable for multiple post-disaster environments is also a challenging problem.In the communication model between UAV-BS and UE,factors such as terrain and buildings in different environments such as villages,suburbs and cities will have different effects on transmission loss.Therefore,it is very necessary to construct a communication model suitable for multiple post-disaster environments.At the same time,in order to equalize the transmission energy consumption for each UAV-BS and serve more UEs,this paper not only maximizes the number of UEs covered by all UAV-BS,but also distributes the number of UEs covered by UAV-BS equally,which is load balancing.In response to the above problems,this paper proposes a post-disaster UAV base station deployment method based on swarm intelligence.First,this paper proposes a post-disaster UAV-BS deployment optimization method based on the UAV-artificial bee colony(U-ABC)algorithm to solve the problem of how to maximize the network throughput by deploying UAV-BS reasonably under the premise of ensuring the communication QoS.Secondly,this paper proposes a post-disaster UAV-BS deployment optimization method based on the global optimal artificial bee colony(GOABC)algorithm to solve the problem of how to maximize the network throughput and the number of UEs covered by UAV-BS through the reasonable deployment of UAV-BS under the premise of ensuring communication QoS in the post-disaster scenario for D2D-CDN.Then,this paper proposes a UAV-BS deployment optimization algorithm based on multi-agent deep deterministic policy gradient(MADDPG)to solve the problem of how to maximize the number of UEs covered by all UAV-BS and take into account the load balance of UAV-BS by reasonably deploying UAV-BS under the premise of ensuring communication QoS in scenarios that are suitable for various post-disaster environments.Finally,through a series of simulation experiments,the effectiveness and advantages of the three proposed UAV-BS deployment optimization methods are verified.The innovative research results achieved in this paper are summarized as follows:(1)Aiming at the problem of UAV-BS deployment optimization in post-disaster scenarios,this paper proposes a post-disaster UAV-BS deployment optimization method based on the U-ABC algorithm.Under the premise of ensuring communication QoS,UAV-BS is deployed reasonably to maximize network throughput.First,this paper proposes an optimization model for UAV-BS deployment after a disaster.Secondly,this paper proposes a U-ABC algorithm to maximize network throughput and calculate the optimal flight position of each UAV-BS.Finally,this paper builds an analysis and test platform.The proposed post-disaster UAV-BS deployment optimization model is analyzed and the effectiveness of the proposed U-ABC algorithm is verified.The results show that the post-disaster UAV-BS deployment optimization method based on the U-ABC algorithm can effectively improve the network throughput and achieve a higher UE coverage.(2)Aiming at the problem of UAV-BS deployment optimization for D2D-CDN post-disaster scenarios,this paper proposes a post-disaster UAV-BS deployment optimization method based on GOABC algorithm.Under the premise of ensuring communication QoS,UAV-BS is deployed reasonably to maximize network throughput and the number of UEs covered by UAV-BS.First,this paper proposes a UAV-BS deployment optimization model U-POM for D2D-CDN after a disaster.Secondly,it is proved that the maximization problem in this paper is an NP-hard problem.The GOABC algorithm is proposed to maximize the network throughput and the number of UEs covered by UAV-BS.Finally,an analysis and test platform was constructed.The proposed U-POM for D2D-CDN is analyzed.The accuracy and stability of the GOABC algorithm are verified.The results show that the post-disaster UAV-BS deployment optimization method based on GOABC algorithm can effectively improve the network throughput and the number of UEs covered by UAV-BS.(3)Aiming at the problem of UAV-BS deployment optimization for multiple post-disaster environment scenarios,this paper proposes a post-disaster UAV-BS deployment optimization method based on the MADDPG algorithm.Under the premise of guaranteeing communication QoS,the reasonable deployment of UAV-BS is used to maximize the number of UEs covered by all UAV-BS and take into account the load balance of UAV-BS.First,this paper proposes a UAV-BS deployment optimization model that is suitable for multiple post-disaster environments.Secondly,this paper proposes a multi-UAV-BS cooperative MADDPG algorithm to maximize the number of UEs covered by UAV-BSs and taking into account load balancing in the communication model.Finally,this paper constructs an analysis and test platform.The proposed UAV-BS deployment optimization model suitable for multiple post-disaster environments is analyzed.The effectiveness of the proposed multi-UAV-BS collaboration MADDPG algorithm is verified.The results show that the post-disaster UAV-BS deployment optimization method based on the MADDPG algorithm can effectively increase the number of UEs covered by UAV-BS and take into account UAV-BS load balancing.